testing a 13E1

[ErikdeBest]

Hi all

Some days ago I bought some 13E1 tubes under the condition that I could test them and in case they were bad I could return them. I do not think they are bad, as they are in white boxes marked which tube they contain (so, no generic white box), and according to the seller, came straight from the Dutch army (code 5960). There are not any spots or whatever on them.

So I want to test them. I set up a test jig, as shown on the picture. My idea is to heat the valves, apply a voltage of -60V on the grid, and slowly, with a Variac, rise the B+ voltage up to about 300V, where, according to the datasheet, the tube should be drawing 200mA.

Today I concluded the test jig, fit in the first 13E1, let it heat up for 30 minutes, and then slowly started to increase the B+. When I arrived at about 180VDC, there was no current flowing, but then, suddenly, things started to go fast with 200V on the plate and -60VDC bias I was getting about 1.5VDC across the 3.9R resistor in the cathode, which corresponds to about 400mA being drawn through the tube. Well, I expected the tube to draw some current, but that much... the tube also made a 'purple' flash... I did not give up I:
- increased bias (turned it more negative)
- current drawn went down
- increased B+
- current drawn went up
- increased bias (turned it more negative)
- current went down..
and so until arriving at -90VDC. But at this point the B+ was maybe at about 280VDC, with 400mA flowing. I was surprised that really small changes in B+ have such influence on current drawn: I know this is common for a tube like 13E1, just like it is for a 6AS7 or 6C33C, but so much... still lots of things to learn.

So, what I ask you is to give a look at the picture, comment on the procedure, etc, as I believe tere is a serious mistake somewhere, and I do not want to lose any 13E1 (although I am sure they can stand some abuse, maybe I am to abusive with them). In the evening I will do my homework, checking the whole circuit again

As to the pin 6: it is the heaters CT, I thought it would form a nice 'star ground', and now the heaters aren't floating anymore. I do not see any problem with it for those test purposes, but if you say: that is complete BS you did there, I will change it.

The HV trafo has an ouput of 330VAC (for 220VAC in), but with the variac any voltage can be selected. The capacitor is a really large 100muF/750VDC MKP.

I thank you all for your patience and time!

Erik de Best

[SY]

Just a guess, but could it be that it's breaking into oscillation? Try it with a 1k stopper, the 100k feed resistor from the bias dropped to 20k, and a small cap (like 0u01) connected from grid to cathode, along with the local bypassing of the plate which you've already done.

I think the connections are right, but your numbering isn't.

[ErikdeBest]

Hi Stuart

Many thanks for replying. I tested your recommendation, but still got weird results. At the end, when I turned down the Variac, I observed that voltage on the cap decreased very slowly...I measured the time, with and without tune connected, and they were the same. Well, tube was not even drawing current... but I fried three resistors (the 3R9 ones, they measured in the M ohm range afterwards).

then I disconnected everything, got a (measured) 98R resistor and put it from cathode to ground, and a 1k resistor from grid to ground. Warmed the valve, and applied B+ until reaching 200V across (anode to cathode) the valve. This happened at 227V. Cathode at 27V, with a 98R resistor = 275mA being drawn. Looking at the curves in MJ's book it seems that one would need less than 200VDC to get this current flowing, that is, the tube should be weak. On the net I found another set of curves, but for short duration pulses. I draw the 270mA line, and where it crosses the 200VDC line it reads that the grid should be at about -35V. My conclusion, again, is that the tube is weak, but I do not know if it applies here. Any comment on this procedure and results is more than welcome.

The first tube I tested, which was the same as the one from the whole procedure from yesterday, had a blue glow between anode and cathode. So I had to check on that also, but seems to be fine. Two other units I tested until now do not show this behavior. I still have some to test!

About the numbering, I have seen that there are differences in numbering of the pins. Duncanamps calls the thick pin number 1
http://tdsl.duncanamps.com/show.php?des=13E1 while on the 13E1 datasheet (also from duncanamps) it is called number 4. I checked and rechecked things about 5 times, but with the results of today I know that at least the connections are right.

[SY]

Blue glow between anode and cathode might be gas- it would be interesting to check grid current. Also, I saw that Morgan Jones had found at least one with a cathode-heater hot spot (therre's a picture of it in one of his books), so I'd keep an eye open for that defect.

It looks like the second tube you tested is close to normal. The differences you're seeing are well within what I'd expect for production spreads. Let's see how the rest of the bunch test out.

[ErikdeBest]

Hi Stuart

I tested some more units, always putting 200VDC across the tube, and the results are pretty much the same.

I more than fast went to test the offender, and it's grid is really dragging current, value is 'proportional' to the B+, that is, as I increased B+, current drawn went up to. At about 230VDC it was about 0.6mA (600mV across 1k GS). I still have some to measure, and I will therefore put my last DVM across the grid stopper, to check if those drag current to. I do not expect so, but, let's see.

I also looked for hot spots when I warmed the tubes up for half an hour, as recommended by MJ. Haven't seen one hotspot until now!

Erik

[SY]

So if you only reject one tube out of that lot, you've still got the bargain of the century.

I hate you.

[ErikdeBest]

Hi Stuart

Can grid current be caused by 'misbehaviour' to the tube, I mean, some thing I did wrong with the former test jig? Sure I can blow up a tube, but this unit is still conducing according to the specs, just that it is pulling some current is not right...

Anyway, you are right, it is a bargain. And, just to make you hate me more, Friday I am going to the same guy to buy the 12 pieces left. The tubes were announced at a 'local ebay', but the seller just received one reaction, and that was mine.

Many thanks for helping me out

Erik

[SY]

Grid current is probably not your fault unless you did something to break the glass-pin seal.

[ErikdeBest]

Thanks Stuart

[ErikdeBest]

I came across the following septar sockets at ePay. Shall they work better than (NOS) russian sockets? I still need good sockets for the +100W dissipation of the 13E1 tubes.

http://cgi.ebay.com/7-Pin-Septar-Tef...QQcmdZViewItem

Thanks, Erik